Next-generation neural interface: ALS patient transmits nearly 2 million words using the power of thought
The science of brain-computer interfaces has achieved a breakthrough, moving from laboratories into real life. Researchers from the University of California, Davis, have presented the results of long-term home use of a speech neuroprosthesis in a patient with amyotrophic lateral sclerosis (ALS). Over 19 months, the patient, named Casey Harrell, transmitted 183,060 sentences, equivalent to 1,960,163 words, at an average speed of 56 words per minute.
The key difference in this work is not a laboratory demonstration, but the everyday use of the system in the absence of researchers. Harrell used the neuroprosthesis for more than 3,800 hours. The system became his voice for communicating with family, friends, colleagues, and doctors. He sent messages, participated in video calls, used the internet, and maintained full-time employment despite complete paralysis.
Accuracy Exceeding Expectations
According to the patient's own assessment, 92% of sentences were decoded "at least mostly correctly." In formal tests, accuracy exceeded 99% with a vocabulary of 125,000 English words, reaching a peak of 99.2%. After assistants were allowed to independently connect the equipment, the average daily interaction time with the system increased from 3.7 to 9.5 hours. As of the publication of the study, Harrell had used the neuroprosthesis on 444 out of 653 days post-implantation.
How It Works
In 2023, the patient had four microelectrode arrays implanted in the left precentral gyrus — the area responsible for coordinating speech. The system reads signals from 256 cortical electrodes when the person attempts to speak. An algorithm converts neural activity into phoneme probabilities every 80 milliseconds, and a language model selects the most likely sequence of words from a vocabulary of 125,000 units. The text is displayed on a screen in real-time, and after the phrase is completed, the system can vocalize it using a synthesized voice, tuned to the patient's voice before the illness.
Notably, the same cursor decoder was used for computer control. It was previously believed that different brain areas were required for speech and movement, but the team proved the opposite: both modes are realized through signals from the speech motor cortex.
From Lab to Home
Initial results were impressive: in 2024, the system achieved 99.6% accuracy with a 50-word vocabulary after a 30-minute training session. The new work shifts the focus to long-term, everyday application. According to the authors, this is one of the key steps toward practical neuroprostheses for people with severe motor impairments. The 3,800 hours of recorded brain activity constitute the largest individual dataset with single-neuron resolution.
However, the system remains experimental. It uses wired connections, requires daily setup by trained assistants, and due to its size, is only suitable for home use. The study describes a single clinical case, and it is not yet known how applicable the results are to other patients and types of implantation.
My comment: This case demonstrates that brain-computer interfaces are ceasing to be a futuristic concept and are becoming a real tool for restoring communication. However, the path to widespread adoption will require miniaturization of equipment and increased system autonomy. Nevertheless, 2 million words transmitted by the power of thought is not just a statistic, but proof that the boundaries between brain and machine are blurring faster than we are used to thinking.